一句话：提出融合三种量子分类器的堆叠集成框架，在ICS电力数据上以少量量子比特实现更优攻击检测。
针对工业控制系统（ICS）网络威胁日益复杂、传统安全方案难以应对的问题，本文探索量子计算在ICS安全检测中的应用。作者提出一种高级堆叠集成框架，集成量子变分分类器（QVC）、量子支持向量分类器（QSVC）和量子神经网络（QNN），在优化量子比特效率的同时提升分类性能。评估使用包含78,345个自然与攻击样本的实时不平衡ICS电力系统数据集。结果表明，该集成框架优于各基线量子分类器，并且仅需少量量子比特即可取得优异结果，展示了量子计算增强ICS网络安全的实际可行性。
The escalating complexity and sophistication of cyber threats targeting Industrial Control Systems (ICS) demand innovative solutions to ensure robust security. Quantum computing, with its unprecedented computational capabilities, emerges as a promising technology in this domain. In this paper, we present an advanced stacking ensemble framework that integrates three quantum classifiers: the Quantum Variational Classifier (QVC), the Quantum Support Vector Classifier (QSVC), and the Quantum Neural Network (QNN). Our ensemble approach optimizes qubit efficiency while delivering superior classification performance. We evaluated our approach on a real-time, imbalanced ICS power system dataset comprising 78, 345 samples representing natural and attack scenarios. The results demonstrate that our advanced ensemble framework outperforms baseline quantum classifiers and achieves these results with minimal qubit usage. This demonstrates our approach's practical viability in enhancing cybersecurity in ICS through quantum computing.
@inproceedings{paperbot9051,
title = {Exploring Advanced Quantum Ensemble for Industrial Control Systems Security},
author = {Danish Vasan and Mohammad Shameem and Mohammad Hammoudeh and Mussadiq Abdul Rahim and Hamad Naeem and Adel F. Ahmed},
booktitle = {FSE},
year = {2025},
doi = {10.1145/3696630.3731620}
}